© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 111617
    Prev
    Next
    INDEX
    Explanations

    Code/HTML snippets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /ay
    -0.29
    ilden
    -0.28
    iment
    -0.27
     perch
    -0.26
    _visibility
    -0.25
     zob
    -0.24
     portrays
    -0.24
    incy
    -0.24
    sut
    -0.24
     -,
    -0.24
    POSITIVE LOGITS
    Ãłng
    0.28
     HAR
    0.28
    棺
    0.28
    ietf
    0.25
    artner
    0.25
     ÑģпÑĢав
    0.24
    åĩºåı£
    0.24
    宣è¨Ģ
    0.24
    riend
    0.24
     plain
    0.24
    Activations Density 0.025%

    No Known Activations